Genetic algorithm applied to simultaneous parameter estimation in bacterial growth.

Several mathematical models have been developed to understand the interactions of microorganisms in foods and predict their growth. The resulting model equations for the growth of interacting cells include several parameters that must be determined for the specific conditions to be modeled. In this study, these parameters were determined by using inverse engineering and a multi-objective optimization procedure that allows fitting more than one experimental growth curve simultaneously. A genetic algorithm was applied to obtain the best parameter values of a model that permit the construction of the front of Pareto with 50 individuals or phenotypes. The method was applied to three experimental data sets of simultaneous growth of lactic acid bacteria (LAB) and Listeria monocytogenes (LM). Then, the proposed method was compared with a conventional mono-objective sequential fit. We concluded that the multi-objective fit by the genetic algorithm gives superior results with more parameter identifiability than the conventional sequential approach.

Mechanistically Inspired Kinetic Approach to Describe Interactions During Co-Culture Growth of Carnobacterium maltaromaticum and Listeria monocytogenes.

Lactic acid bacteria and Listeria monocytogenes are psychotropic organisms that can grow and compete in food such as lightly preserved fishery products. Predictive microbiology is nowadays one of the leading tools to assess the behavior of bacteria in food and to predict food spoilage. Mathematical models can be used to predict the growth, inactivation or growth probability of bacteria. Currently, the efforts in microbial modeling are oriented towards extrapolation of results beyond experiments in order to predict the growth of interacting microorganisms and develop new food preservation processes. In the present work, a model combining both heterogeneous population and quasi-chemical approaches to describe the different phases of the bacterial growth curve is presented. The model was applied to both monoculture and co-culture cases of lactic acid bacteria, Carnobacterium maltaromaticum H-17, and two Listeria monocytogenes strains in a raw fish extract. It is a highlight that our model includes novel inhibition reactions due to the accumulation of metabolites, and a general equation to take into account the effect of chemical compounds during the lag or physiological adaptation phase of the cells. Our results show that the proposed model can accurately describe the experimental data when the curve shape is a sigmoid, and when it presents a maximum. Besides, the parameters have biological interpretability since the model is mechanistically inspired.